We began this project with a study of squamous cell carcinoma of the oral cavity and epidermis in humans, finding defective differentiation in these cells. Our identification of the 40 kilodalton keratin in SCC cells then led us to study other epithelial cell types in culture and disclosed the expression of: (1)\cell-type-specific subsets of a large family of keratins and (2)\cell-type-specific growth requirements by mesothelial cells, urothelial cells, and kidney tubule epithelial cells. In addition to defining culture conditions for long-term, clonal growth of these cell types, we have begun to study the nature of dedifferentiation in mesoderm-derived epithelial cell types. We have found that normal mesothelial cells dedifferentiate reversibly during rapid growth in culture and that this process, which includes a morphological change to fibroblastoid appearance and the expression of vimentin intermediate filaments, occurs also in mesothelioma. We are now seeking to identify and characterize the various cell types that grow in culture from normal human kidney cortex. We have also substantially improved conditions for keratinocyte growth in culture and have optimized conditions for identifying and isolating rare, induced mutants from these feeder layerdependent cells, thus setting the stage for seeking chemical-induced transformation events in cultured populations of normal human keratinocytes. (M)